ihanif's picture
update model card README.md
1be152f
|
raw
history blame
2.39 kB
metadata
license: apache-2.0
tags:
  - whisper-event
  - generated_from_trainer
datasets:
  - google/fleurs
metrics:
  - wer
model-index:
  - name: Whisper Medium Pashto
    results:
      - task:
          name: Automatic Speech Recognition
          type: automatic-speech-recognition
        dataset:
          name: google/fleurs ps_af
          type: google/fleurs
          config: ps_af
          split: test
          args: ps_af
        metrics:
          - name: Wer
            type: wer
            value: 50.56749394673123

Whisper Medium Pashto

This model is a fine-tuned version of openai/whisper-medium on the google/fleurs ps_af dataset. It achieves the following results on the evaluation set:

  • Loss: 1.4603
  • Wer: 50.5675

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

  • learning_rate: 3e-07
  • train_batch_size: 32
  • eval_batch_size: 16
  • seed: 42
  • gradient_accumulation_steps: 2
  • total_train_batch_size: 64
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 50
  • training_steps: 1000
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Wer
0.0334 14.29 100 1.0348 50.0908
0.0021 28.57 200 1.1971 49.4855
0.0007 42.86 300 1.2651 49.7352
0.0006 57.14 400 1.3084 49.9697
0.0005 71.43 500 1.3479 50.0605
0.0004 85.71 600 1.3835 50.3027
0.0004 100.0 700 1.4139 50.4540
0.0004 114.29 800 1.4382 50.4616
0.0004 128.57 900 1.4545 50.5297
0.0003 142.86 1000 1.4603 50.5675

Framework versions

  • Transformers 4.26.0.dev0
  • Pytorch 1.13.0+cu117
  • Datasets 2.7.1.dev0
  • Tokenizers 0.13.2